BookmarkSubscribeRSS Feed
stoffprof
Calcite | Level 5

This question is related to a question I posted in the SAS/STAT forum ().

I have a large and sparse design matrix of ones and zeros (about 450k rows, 2500 columns). I'm trying to use this to run a regression, but SAS/STAT procedures are taking too long. (I ran proc hpmixed and it had not produced any results after 24 hours.)

If I read the data into Matlab as a sparse matrix, I can run a QR decomposition and invert the resulting matrix in less than 10 seconds. Is it possible to take this approach in IML? I feel like I must be missing something because it doesn't seem like this is a hard problem to solve, so I thought SAS would handle everything on the fly, but it seems that I'm going wrong somewhere.

Thanks for any advice.

1 REPLY 1
Rick_SAS
SAS Super FREQ

In your MATLAB example, if X = QR, then R and Q are dense 2500x2500 matrices, So I guess you are asking whether IML has a sparse QR decomposition?  No, it does not.

sas-innovate-2024.png

Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.

Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.

 

Register now!

Multiple Linear Regression in SAS

Learn how to run multiple linear regression models with and without interactions, presented by SAS user Alex Chaplin.

Find more tutorials on the SAS Users YouTube channel.

From The DO Loop
Want more? Visit our blog for more articles like these.
Discussion stats
  • 1 reply
  • 895 views
  • 0 likes
  • 2 in conversation